HNHacker News
TopNewBestAskShowJobs

rphlx

1,109 karma · joined October 8, 2009

submissionscomments
rphlx··on DOD Just Beginning to Grapple with Scale of Weapon Systems Vulnerabilities
A decent fraction of the population views password restrictions as a challenge to come up with the shittiest, least secure password that they possibly can while still meeting all restrictions. You can blame users for that with some justice, but as a system designer, it's still your responsibility provide security despite shitty but reasonably likely human behavior.

With modern crypto there are very few systems where it's appropriate to have a user-selected <=12 character password for primary auth, yet unfortunately that continues to be widespread for banks, ecommerce, and (probably) some military systems. High-end security people seem to almost universally hate short user-selected passwords (except when they have to break them..) but old practices die hard and old systems take a long time to be replaced.

rphlx··on Bloomberg Reports China Infiltrated the Supermicro Supply Chain We Investigate
It's possible that some servers hang the BIOS flash off the BMC, or (more likely) at least have some way for the BMC to write to it - if not by design, then through a HW vuln further up the stack.

And I believe that once you control the BIOS image you control the boot chain of trust.

rphlx··on Bloomberg Reports China Infiltrated the Supermicro Supply Chain We Investigate
> Bloomberg says it is in line with memory to CPUs to intercept some password validation code

I think that's a misreading of their article. They were not claiming that's what was actually done, they just provided that as an example of what a HW attacker could do. Later on I remember them saying that the malicious part was connected to the BMC, not the main CPU. If there's a serious USB vuln in the BMC, then four wires could be enough to compromise it and gain God Mode over the early x86 SW environment.

rphlx··on Show HN: A new protocol for packet radio including ECDSA and compression
FWIW there's no conflict there in terms of the cryptography.

A replay-protected, keyed authenticator of plaintext commands (HMAC, etc) is not encryption; it is authentication. In a similar sense you are allowed to put a rolling log-in password on a packet radio BBS, so that passive monitoring will see the password, but as soon as it's revealed, it's no longer useful for additional logins.

rphlx··on Modifications to Google Chromium for removing Google integration
I completely agree. The rapid flux in these huge browser projects makes personal security audits impossible even if you are one of the top 10 super-rigorous C++ devs in the world. Would be nice if some project forked the browser of their choice and then got some very careful well-respected teams/rockstars to audit it, and from then on, applied only security fixes and outright-rejected dubious stuff like WebUSB, google auto-login, etc. With Chrome and Chromium, sure, you get a very good security team working on it for you at Google, but you also get changes which are not in your best interests and/or undermine security directly or indirectly like, well, WebUSB. This is not ideal.
rphlx··on Google’s first all-hands after 2016 election [video]
Google appears to have a long history of infantilizing its workforce, partly - one hopes - as a sort of group-bonding in-joke, and partly - one suspects - as a darker psychological management strategy with objectives that are not always unambiguously in the infant's best interest.
rphlx··on Google’s first all-hands after 2016 election [video]
At best a crackpot one with no serious support from contemporary leaders such as Churchill, or historians since then AFAIK. Putting the war in the Pacific aside, I think you can place much of the blame for WWII on the election of an extremist in Germany, and that had little to do with any alleged "boredom" in the electorate. A fairly decent case may be made for "anger", "disillusion", "jealousy" and "resentment" though, and I suppose you could make a case for those also being important factors in the 2016 US election.
rphlx··on Google’s first all-hands after 2016 election [video]
The view that someone is automatically a monster or an "oppressor" because they voted Republican is wrong and offensive in the extreme.

It is possible to vote Republican without wholly supporting - let alone personally imitating - everything that Trump does, just as it is possible to vote Democrat without wholly supporting everything Obama did.

You vote for the party that gives you the most of what you want and has a reasonable chance of being elected. Rarely if ever will you be entirely satisfied with their personal behavior or policies once in office.

rphlx··on Google’s first all-hands after 2016 election [video]
Be wary of the legal implications. At minimum GOOG can probably claim copyright and do a DMCA takedown.
rphlx··on Google’s first all-hands after 2016 election [video]
If only it were so simple. What you are proposing seems to require universal acceptance of an authoritarian regime that sets the social rules which are, as you say, "not negotiable". Otherwise in a free society almost everything is negotiable and a precise definition of common decency is not so common, particularly across cultures. Similarly, trying to produce a complete and widely-accepted definition of intolerance is almost certain to elicit plenty of actual intolerance, but not enough to actually compile a complete and widely-accepted definition of intolerance.
rphlx··on Intel can’t supply 14nm Xeons, HPE recommends AMD Epyc
Intel's recovery a decade ago also came down to superior mfg, and that past strength is their main weakness this time around. I am very concerned about their ability to yield 24+ core monolithic server parts on 10nm at competitive cost/perf/timeframe, given that 10nm seems to be an out-of-control dumpster fire even compared to their 14nm node which was itself around 2 years late.
rphlx··on Intel can’t supply 14nm Xeons, HPE recommends AMD Epyc
I like RISC-V more than is probably healthy, but it is not a player at all, let alone a compelling player, vs x86_64 or aarch64 in the server space and probably won't be for at least 2-3 years. One obvious missing feature is robust KVM/type 2 hyp support.
rphlx··on Intel can’t supply 14nm Xeons, HPE recommends AMD Epyc
There's a cheap and timeless emotional thrill in supporting a rising underdog, but if Intel declines too far, that thrill will not be cheap, but very expensive for most of us here. AMD will gladly collect the same quasi-monopoly margins that Intel enjoyed for the past decade, should Intel become enough of a hot mess to permit that. So be a bit careful what you wish for.
rphlx··on Broken Genius – the case of William Shockley (2013)
If the government offers you a rebate of prior taxes paid, it's rational and in your self interest to accept it, even if you do not think the program providing it should exist.
rphlx··on Broken Genius – the case of William Shockley (2013)
I certainly agree with the top half of your post but I believe the bottom is mischaracterizing Rand. As far as I know she did not wish to form a violent authoritarian state a la Lenin to "plan" society into a certain pattern. She seemed to think her desires would, when enough constraints on reasoned self-interest were removed, realize themselves through individual action. And she certainly did not forget that individuals have agency; on the contrary, Objectivism relies on individual self-interest to the intentional exclusion of pretty much all else.
rphlx··on Broken Genius – the case of William Shockley (2013)
> Who you are is almost entirely contained within your genes, with a minority contribution from enviroment and parents and schooling. Nature won the debate

It would be more precise to say that maximum potential is largely genetic. Watson was wrong in that sense, but right in the opposite sense: an extremely bad environment - malnutrition, disease, pollution, very poor or no education, and so on - can definitely prevent someone from reaching their genetic potential. Unfortunately, an extremely good environment does not seem to reliably eliminate a genetic ceiling (for want of a better term) to a similar extent.

rphlx··on How LLVM Optimizes a Function
According to Compiler Explorer, while 6.0.0 does it, llvm trunk is no longer doing the GVN load optimization. It reverted to using a double load:

        mov     edx, dword ptr [rdi + 4*rcx]
        cmp     edx, dword ptr [rdi + 4*rcx + 4]
per loop. Dunno if intentional or just a regression.
rphlx··on Unveil(2) – Unveil parts of a restricted filesystem view
This is neat but I am still partial to slower but fully generic approaches for syscall filtering, such as BPF.

Sooner or later you run out of clever English verbs for the plethora of fixed-function ones.

rphlx··on Djbsort: A new software library for sorting arrays of integers
It is sadly necessary; 30%+ performance regressions from, say, gcc 4 to gcc 6 are not uncommon w/ vector intrinsics.
rphlx··on Djbsort: A new software library for sorting arrays of integers
FWIW Haswell - on which this was benchmarked - does not downclock on AVX2; it just boosts vcore and thermal throttles if the cooling solution can't keep the die below ~100C (which of course, the crappy TIM plus bundled desktop heatsink have essentially zero chance of doing with many extended workloads). To preclude thermal throttling the auto-downclock-on-AVX2 was added later, I think for Skylake.
rphlx··on Djbsort: A new software library for sorting arrays of integers
Transferring SIMD/FPU regs to/from memory is actually fast - a modest fraction of the typical context switch cost - on modern x86_64 thanks to XSAVE.
rphlx··on iOS 11.4 to Disable USB Port After 7 Days: What It Means for Mobile Forensics
In general HN has a strong bias toward Apple over Android RE: security. Many of the points are entirely valid, but it's also true that some important advantages on the Android side (such as those listed above) as often understated or overlooked here.

For me, a closed-source OS is a dealbreaker on its own, regardless of any other major HW or SW advantages.

rphlx··on iOS 11.4 to Disable USB Port After 7 Days: What It Means for Mobile Forensics
In most circumstances it's not clear to me that a >48 hour timeout is really a major impediment to law enforcement in a major metro. Of course, if they arrest like 10,000 people on the same day from a major protest or something it could create a capacity problem, but otherwise it seems like more of an irritant - a new time constraint - rather than a truly effective countermeasure to vulns in their LI/USB stack.

I would be much more impressed if they allowed users to wholesale disable all non-charging functionality.

rphlx··on Tell HN: Sci-Hub's TLS certificate has started failing
> Microsoft's only option is to completely drop the root cert, right? So there's no real non-nuclear option...

In small-scale disputes MS (and other browser vendors) would not have to nuke an entire large CA to get their way. In principle they could just blacklist the individual certs/names, leaving the CA's other certs alone.

That ability/implied threat probably does mean that the CAs tend to comply with MS piracy/copyright-related revocation requests, because refusing to comply would piss off MS (and possibly law enforcement) without actually stopping them from getting their way by other means.

rphlx··on Hijack of Amazon’s domain service used to reroute web traffic for two hours
Not to start another DNSSEC melee (as I start another DNSSEC melee..) but for it to be effective we really need browsers to a) be able to tell whether a DNS response was properly DNSSEC signed or not, b) produce some large, scary, red warning for ones that aren't, mark them "Not Secure", etc. [1]

There is a major chicken-and-egg/game-theoretical problem though: any browser that does that today will piss off/irritate its users, forcing them to use other browsers or older versions of the same browser, as DNSSEC is not widely deployed on the corporate side. And until most/all browsers do something major to make the current DNS security crisis obvious to large numbers of users, most companies won't care enough to deploy DNSSEC.

At least it will be interesting to see whether, and how, this eventually gets solved.

[1] Certain abysmal DNSSEC cryptographic choices, such as 512-1024-bit RSA, should also be addressed.

rphlx··on Cryptographic Right Answers
> Avoid: the OpenSSL RNG

Is that just historical - i.e. a tiny chance somebody is still using the broken Debian version from a decade ago - or is there actually something still insecure, or at least suspicious, even in 1.1.0+?

I ask only because getentropy() is not widely available in the Linux world yet, and getrandom() - and direct use of the /dev files - do have some caveats of their own.

rphlx··on The End of Windows
Indeed. If/When another major conflict comes to the developed world, you better believe major cloud DCs will be pretty high up on the cyber and (if it comes to it) kinetic target lists. Never have so few facilities been relied upon for so much by so many.
rphlx··on The End of Windows
> Windows as a separate company probably cannot make it long term.

That is surely overstating it, at least for definitions of "long" meaning <= 20 years. If you assume that Windows is more or less "done" their engineering expenses could be low, and revenue significant, merely serving the corporate and various niche (gaming, etc) markets which are still heavily bound to Windows. Their market share is declining in some niches (such as CAD) and maybe even overall (to OS X), but negative revenue and market share growth does not necessarily mean quick extinction. Some dinosaurs take decades to bleed out.

rphlx··on Just one QUIC bit
> the only target audience for that consists of Radar techs walking around in front of hugeass antennas

And perhaps US diplomats in Cuba.

rphlx··on Logs in High Sierra Show Plaintext Password for APFS Encrypted External Volumes
Closed source disk encryption products: Not Even Once.
← PreviousPage 2 of 15Next →